9 files changed, 62 insertions, 48 deletions

diff --git a/net/ipv4/icmp.c b/net/ipv4/icmp.c
index 3d78464f64ea..badfc5849973 100644
--- a/net/ipv4/icmp.c
+++ b/net/ipv4/icmp.c
@@ -349,12 +349,12 @@ static void icmp_push_reply(struct icmp_bxm *icmp_param,
 {
         struct sk_buff *skb;
 
-        ip_append_data(icmp_socket->sk, icmp_glue_bits, icmp_param,
-                       icmp_param->data_len+icmp_param->head_len,
-                       icmp_param->head_len,
-                       ipc, rt, MSG_DONTWAIT);
-
-        if ((skb = skb_peek(&icmp_socket->sk->sk_write_queue)) != NULL) {
+        if (ip_append_data(icmp_socket->sk, icmp_glue_bits, icmp_param,
+                           icmp_param->data_len+icmp_param->head_len,
+                           icmp_param->head_len,
+                           ipc, rt, MSG_DONTWAIT) < 0)
+                ip_flush_pending_frames(icmp_socket->sk);
+        else if ((skb = skb_peek(&icmp_socket->sk->sk_write_queue)) != NULL) {
                 struct icmphdr *icmph = skb->h.icmph;
                 unsigned int csum = 0;
                 struct sk_buff *skb1;
diff --git a/net/ipv4/inetpeer.c b/net/ipv4/inetpeer.c
index 95473953c406..ab18a853d7ce 100644
--- a/net/ipv4/inetpeer.c
+++ b/net/ipv4/inetpeer.c
@@ -450,10 +450,13 @@ static void peer_check_expire(unsigned long dummy)
         /* Trigger the timer after inet_peer_gc_mintime .. inet_peer_gc_maxtime
          * interval depending on the total number of entries (more entries,
          * less interval). */
-        peer_periodic_timer.expires = jiffies
-                + inet_peer_gc_maxtime
-                - (inet_peer_gc_maxtime - inet_peer_gc_mintime) / HZ *
-                        peer_total / inet_peer_threshold * HZ;
+        if (peer_total >= inet_peer_threshold)
+                peer_periodic_timer.expires = jiffies + inet_peer_gc_mintime;
+        else
+                peer_periodic_timer.expires = jiffies
+                        + inet_peer_gc_maxtime
+                        - (inet_peer_gc_maxtime - inet_peer_gc_mintime) / HZ *
+                                peer_total / inet_peer_threshold * HZ;
         add_timer(&peer_periodic_timer);
 }
 
diff --git a/net/ipv4/ipcomp.c b/net/ipv4/ipcomp.c
index 2065944fd9e5..7ded6e60f43a 100644
--- a/net/ipv4/ipcomp.c
+++ b/net/ipv4/ipcomp.c
@@ -358,7 +358,7 @@ static struct crypto_tfm **ipcomp_alloc_tfms(const char *alg_name)
         int cpu;
 
         /* This can be any valid CPU ID so we don't need locking. */
-        cpu = smp_processor_id();
+        cpu = raw_smp_processor_id();
 
         list_for_each_entry(pos, &ipcomp_tfms_list, list) {
                 struct crypto_tfm *tfm;
diff --git a/net/ipv4/netfilter/ip_queue.c b/net/ipv4/netfilter/ip_queue.c
index eda1fba431a4..c6baa8174389 100644
--- a/net/ipv4/netfilter/ip_queue.c
+++ b/net/ipv4/netfilter/ip_queue.c
@@ -214,6 +214,12 @@ ipq_build_packet_message(struct ipq_queue_entry *entry, int *errp)
                 break;
         
         case IPQ_COPY_PACKET:
+                if (entry->skb->ip_summed == CHECKSUM_HW &&
+                    (*errp = skb_checksum_help(entry->skb,
+                                               entry->info->outdev == NULL))) {
+                        read_unlock_bh(&queue_lock);
+                        return NULL;
+                }
                 if (copy_range == 0 || copy_range > entry->skb->len)
                         data_len = entry->skb->len;
                 else
@@ -385,6 +391,7 @@ ipq_mangle_ipv4(ipq_verdict_msg_t *v, struct ipq_queue_entry *e)
         if (!skb_ip_make_writable(&e->skb, v->data_len))
                 return -ENOMEM;
         memcpy(e->skb->data, v->payload, v->data_len);
+        e->skb->ip_summed = CHECKSUM_NONE;
         e->skb->nfcache |= NFC_ALTERED;
 
         /*
diff --git a/net/ipv4/netfilter/ipt_ECN.c b/net/ipv4/netfilter/ipt_ECN.c
index ada9911118e9..94a0ce1c1c9d 100644
--- a/net/ipv4/netfilter/ipt_ECN.c
+++ b/net/ipv4/netfilter/ipt_ECN.c
@@ -61,16 +61,20 @@ set_ect_tcp(struct sk_buff **pskb, const struct ipt_ECN_info *einfo, int inward)
         if (!tcph)
                 return 0;
 
-        if (!(einfo->operation & IPT_ECN_OP_SET_ECE
-              || tcph->ece == einfo->proto.tcp.ece)
-            && (!(einfo->operation & IPT_ECN_OP_SET_CWR
-                  || tcph->cwr == einfo->proto.tcp.cwr)))
+        if ((!(einfo->operation & IPT_ECN_OP_SET_ECE) ||
+             tcph->ece == einfo->proto.tcp.ece) &&
+            ((!(einfo->operation & IPT_ECN_OP_SET_CWR) ||
+             tcph->cwr == einfo->proto.tcp.cwr)))
                 return 1;
 
         if (!skb_ip_make_writable(pskb, (*pskb)->nh.iph->ihl*4+sizeof(*tcph)))
                 return 0;
         tcph = (void *)(*pskb)->nh.iph + (*pskb)->nh.iph->ihl*4;
 
+        if ((*pskb)->ip_summed == CHECKSUM_HW &&
+            skb_checksum_help(*pskb, inward))
+                return 0;
+
         diffs[0] = ((u_int16_t *)tcph)[6];
         if (einfo->operation & IPT_ECN_OP_SET_ECE)
                 tcph->ece = einfo->proto.tcp.ece;
@@ -79,13 +83,10 @@ set_ect_tcp(struct sk_buff **pskb, const struct ipt_ECN_info *einfo, int inward)
         diffs[1] = ((u_int16_t *)tcph)[6];
         diffs[0] = diffs[0] ^ 0xFFFF;
 
-        if ((*pskb)->ip_summed != CHECKSUM_HW)
+        if ((*pskb)->ip_summed != CHECKSUM_UNNECESSARY)
                 tcph->check = csum_fold(csum_partial((char *)diffs,
                                                      sizeof(diffs),
                                                      tcph->check^0xFFFF));
-        else
-                if (skb_checksum_help(*pskb, inward))
-                        return 0;
         (*pskb)->nfcache |= NFC_ALTERED;
         return 1;
 }
diff --git a/net/ipv4/netfilter/ipt_TCPMSS.c b/net/ipv4/netfilter/ipt_TCPMSS.c
index 1049050b2bfb..7b84a254440e 100644
--- a/net/ipv4/netfilter/ipt_TCPMSS.c
+++ b/net/ipv4/netfilter/ipt_TCPMSS.c
@@ -61,6 +61,10 @@ ipt_tcpmss_target(struct sk_buff **pskb,
         if (!skb_ip_make_writable(pskb, (*pskb)->len))
                 return NF_DROP;
 
+        if ((*pskb)->ip_summed == CHECKSUM_HW &&
+            skb_checksum_help(*pskb, out == NULL))
+                return NF_DROP;
+
         iph = (*pskb)->nh.iph;
         tcplen = (*pskb)->len - iph->ihl*4;
 
@@ -186,9 +190,6 @@ ipt_tcpmss_target(struct sk_buff **pskb,
                newmss);
 
  retmodified:
-        /* We never hw checksum SYN packets.  */
-        BUG_ON((*pskb)->ip_summed == CHECKSUM_HW);
-
         (*pskb)->nfcache |= NFC_UNKNOWN | NFC_ALTERED;
         return IPT_CONTINUE;
 }
diff --git a/net/ipv4/tcp.c b/net/ipv4/tcp.c
index ddb6ce4ecff2..69b1fcf70077 100644
--- a/net/ipv4/tcp.c
+++ b/net/ipv4/tcp.c
@@ -584,7 +584,7 @@ static inline void skb_entail(struct sock *sk, struct tcp_sock *tp,
         sk_charge_skb(sk, skb);
         if (!sk->sk_send_head)
                 sk->sk_send_head = skb;
-        else if (tp->nonagle&TCP_NAGLE_PUSH)
+        if (tp->nonagle & TCP_NAGLE_PUSH)
                 tp->nonagle &= ~TCP_NAGLE_PUSH; 
 }
 
diff --git a/net/ipv4/tcp_ipv4.c b/net/ipv4/tcp_ipv4.c
index 5d91213d34c0..67c670886c1f 100644
--- a/net/ipv4/tcp_ipv4.c
+++ b/net/ipv4/tcp_ipv4.c
@@ -242,9 +242,14 @@ static int tcp_v4_get_port(struct sock *sk, unsigned short snum)
                 tcp_port_rover = rover;
                 spin_unlock(&tcp_portalloc_lock);
 
-                /* Exhausted local port range during search? */
+                /* Exhausted local port range during search?  It is not
+                 * possible for us to be holding one of the bind hash
+                 * locks if this test triggers, because if 'remaining'
+                 * drops to zero, we broke out of the do/while loop at
+                 * the top level, not from the 'break;' statement.
+                 */
                 ret = 1;
-                if (remaining <= 0)
+                if (unlikely(remaining <= 0))
                         goto fail;
 
                 /* OK, here is the one we will use.  HEAD is
diff --git a/net/ipv4/tcp_output.c b/net/ipv4/tcp_output.c
index 3ed6fc15815b..dd30dd137b74 100644
--- a/net/ipv4/tcp_output.c
+++ b/net/ipv4/tcp_output.c
@@ -861,7 +861,8 @@ static int tso_fragment(struct sock *sk, struct sk_buff *skb, unsigned int len,
         u16 flags;
 
         /* All of a TSO frame must be composed of paged data.  */
-        BUG_ON(skb->len != skb->data_len);
+        if (skb->len != skb->data_len)
+                return tcp_fragment(sk, skb, len, mss_now);
 
         buff = sk_stream_alloc_pskb(sk, 0, 0, GFP_ATOMIC);
         if (unlikely(buff == NULL))
@@ -924,10 +925,6 @@ static int tcp_tso_should_defer(struct sock *sk, struct tcp_sock *tp, struct sk_
 
         limit = min(send_win, cong_win);
 
-        /* If sk_send_head can be sent fully now, just do it.  */
-        if (skb->len <= limit)
-                return 0;
-
         if (sysctl_tcp_tso_win_divisor) {
                 u32 chunk = min(tp->snd_wnd, tp->snd_cwnd * tp->mss_cache);
 
@@ -974,6 +971,8 @@ static int tcp_write_xmit(struct sock *sk, unsigned int mss_now, int nonagle)
 
         sent_pkts = 0;
         while ((skb = sk->sk_send_head)) {
+                unsigned int limit;
+
                 tso_segs = tcp_init_tso_segs(sk, skb, mss_now);
                 BUG_ON(!tso_segs);
 
@@ -994,9 +993,10 @@ static int tcp_write_xmit(struct sock *sk, unsigned int mss_now, int nonagle)
                                 break;
                 }
 
+                limit = mss_now;
                 if (tso_segs > 1) {
-                        u32 limit = tcp_window_allows(tp, skb,
-                                                      mss_now, cwnd_quota);
+                        limit = tcp_window_allows(tp, skb,
+                                                  mss_now, cwnd_quota);
 
                         if (skb->len < limit) {
                                 unsigned int trim = skb->len % mss_now;
@@ -1004,15 +1004,12 @@ static int tcp_write_xmit(struct sock *sk, unsigned int mss_now, int nonagle)
                                 if (trim)
                                         limit = skb->len - trim;
                         }
-                        if (skb->len > limit) {
-                                if (tso_fragment(sk, skb, limit, mss_now))
-                                        break;
-                        }
-                } else if (unlikely(skb->len > mss_now)) {
-                        if (unlikely(tcp_fragment(sk, skb,  mss_now, mss_now)))
-                                break;
                 }
 
+                if (skb->len > limit &&
+                    unlikely(tso_fragment(sk, skb, limit, mss_now)))
+                        break;
+
                 TCP_SKB_CB(skb)->when = tcp_time_stamp;
 
                 if (unlikely(tcp_transmit_skb(sk, skb_clone(skb, GFP_ATOMIC))))
@@ -1064,11 +1061,14 @@ void tcp_push_one(struct sock *sk, unsigned int mss_now)
         cwnd_quota = tcp_snd_test(sk, skb, mss_now, TCP_NAGLE_PUSH);
 
         if (likely(cwnd_quota)) {
+                unsigned int limit;
+
                 BUG_ON(!tso_segs);
 
+                limit = mss_now;
                 if (tso_segs > 1) {
-                        u32 limit = tcp_window_allows(tp, skb,
-                                                      mss_now, cwnd_quota);
+                        limit = tcp_window_allows(tp, skb,
+                                                  mss_now, cwnd_quota);
 
                         if (skb->len < limit) {
                                 unsigned int trim = skb->len % mss_now;
@@ -1076,15 +1076,12 @@ void tcp_push_one(struct sock *sk, unsigned int mss_now)
                                 if (trim)
                                         limit = skb->len - trim;
                         }
-                        if (skb->len > limit) {
-                                if (unlikely(tso_fragment(sk, skb, limit, mss_now)))
-                                        return;
-                        }
-                } else if (unlikely(skb->len > mss_now)) {
-                        if (unlikely(tcp_fragment(sk, skb, mss_now, mss_now)))
-                                return;
                 }
 
+                if (skb->len > limit &&
+                    unlikely(tso_fragment(sk, skb, limit, mss_now)))
+                        return;
+
                 /* Send it out now. */
                 TCP_SKB_CB(skb)->when = tcp_time_stamp;